Content uploaded by Andy Haines
Author content
All content in this area was uploaded by Andy Haines
Content may be subject to copyright.
GENERAL
PRACTICE
Visual
problems
in
the
elderly
population
and
implications
for
services
R
P
L
Wormald,
L
A
Wright,
P
Courtney,
B
Beaumont,
A
P
Haines
Abstract
Objective-To
determine
the
prevalence
of
visual
disability
and
common
eye
disease
among
elderly
people
in
inner
London.
Design-Cross
sectional
random
sample
survey.
Setting-Inner
London
health
centre.
Subjects-Random
sample
of
people
aged
65
and
over
taken
from
practice's
computerised
age-sex
register.
Main
outcome
measures-Presenting
binocular
Snelien
6
m
distance
acuity
and
best
monocular
3
m
Sonksen-Silver
acuity
to
classify
prevalence
of
blind-
ness
by
World
Health
Organisation
criteria
(less
than
3/60
in
better
eye)
and
American
criteria
for
legal
blindness
(better
eye
equal
to
6/60
or
less)
and
of
low
vision
by
WHO
criteria
(best
acuity
6/18)
and
visual
impairment
by
American
criteria
(less
than
6/12
or
20/40
but
greater
than
6/60
or
20/200
in
better
eye).
Principal
cause
of
visual
loss
by
diagnosis,
referral
indication
by
cause
to
hospital
eye
service,
and
proportion
of
cases
known
to
primary
care.
Results-207
of
288
(72%)
eligible
people
were
examined.
17
(8%)
housebound
subjects
were
examined
at
home.
The
prevalence
of
blindness
was
1%
by
WHO
criteria
and
3*9%
by
American
criteria.
The
prevalence
of
low
vision
(WHO
criteria)
was
7-7%.
The
prevalence
of
visual
impairment
(American
criteria)
was
10-6%.
Cataract
accounted
for
75%
of
cases
of
low
vision.
Only
eight
out
of
16
patients
with
low
vision
were
known
by
their
general
practitioner
to
have
an
eye
problem.
56
subjects
(27%)
would
probably
have
benefited
from
refraction.
Comparisons
with
studies
in
the
United
States
and
Finland
suggested
higher
rates
in
this
sample,
mainly
due
to
the
prevalence
of
disabling
cataract.
Conclusion-There
seems
to
be
a
considerable
amount
of
undetected
ocular
disease
in
elderly
people
in
the
community.
Introduction
Epidemiological
data
on
visual
impairment
in
the
United
Kingdom
are
scarce.
The
registers
of
blindness
and
partial
sight
include
data
on
cause
and
minimal
demographic
information
but
neither
can
be
relied
on
for
epidemiological
purposes.'
A
national
household
survey
in
the
1970s
found
that
40%
of
visually
impaired
people
had
never
seen
an
ophthalmologist,2
and
the
Office
of
Population
Censuses
and
Surveys
disability
survey3
estimated
that
about
twice
as
many
people
are
visually
impaired
as
can
be
accounted
for
by
the
registered
population
(blind
and
partially
sighted).
Elderly
people
are
at
the
greatest
risk
of
visual
impairment.
A
community
study
conducted
in
Melton
Mowbray,
Leicestershire,4
found
high
prevalences
of
common
visually
disabling
conditions
in
a
sample
of
patients
aged
75
and
over.
The
Framingham
eye
survey'
represented
mainly
middle
class
white
Americans
aged
over
55
and
is
most
frequently
cited
for
prevalence
estimates
of
eye
disease
in
industrialised
countries.
In
London
there
is
a
range
of
facilities
available
for
ophthalmological
services
but
the
extent
to
which
the
needs
of
elderly
people
are
being
met
is
not
known.
The
number
of
elderly
people-particularly
those
aged
75
and
over-is
expected
to
increase
significantly
in
the
future,
which
will
lead
to
an
increased
burden
of
visual
disability
and
ocular
morbidity.6
The
following
survey
was
undertaken
to
estimate
the
prevalence
of
common
eye
disorders
in
an
elderly
inner
city
population
and
to
see
whether
people
with
visual
impairment
were
known
to
services.
Subjects
and
methods
We
present
the
findings
from
a
50%
random
sample
of
patients
aged
65
and
over
from
a
practice
with
a
total
population
of
around
9000.
Information
on
non-
attenders
in
this
subsample
was
available
from
the
practice
records.
A
total
of
288
subjects
aged
65
and
over
were
identified
from
the
age-sex
register
and
invited
by
letter
from
the
general
practitioners
to
participate
in
a
survey
of
eye
health.
The
survey
clinic
was
established
in
a
day
centre
for
the
elderly
situated
immediately
next
door
to
the
practice
premises.
Refusers
were
asked
to
give
reasons
for
not
wishing
to
participate.
A
brief
questionnaire
asking
for
demographic
and
socioeconomic
details
as well as
relevant
medical
and
ophthalmic
history
preceded
acomplete
ocular
exam-
ination.
Binocular
Snellen
6
m
visual
acuity
with
any
spectacles
normally
worn
and
uniocular
3
m
Sonksen-
Silver
visual
acuity
(measured
with
a
pin
hole
when
6/9
equivalent
was
not
achieved),
aided
by
spectacles
if
usually
worn,
were
taken
as
the
last
line
read
correctly.
Near
vision
was
recorded
as
being
able
to
read
N6
with
normal
reading
correction
as
required.
Improvement
in
distance
acuity
of
greater
than
one
Snellen
(equiva-
lent)
line
with
pin
hole
with
one
or
both
eyes
was
taken
as
an
indication
of
potential
benefit
from
refraction.
Similarly,
poor
near
vision
in
the
presence
of
good
distance
acuity
was
an
indication
of
requirement
for
new
near
correction.
Therefore,
formal
refraction
was
not
undertaken
but
spectacles
were
assessed
for
their
adequacy.
The
central
visual
field
was
tested
with
the
Henson
CFS2000
in
all
participants
attending
the
clinic
who
could
perform
the
test
(8
1%).
Confrontation
fields
were
used
when
appropriate
for
those
unable
to
manage
automated
testing
and
on
home
visits.
Intraocular
pressure
was
tested
with
the
Perkins
mark
2
applana-
tion
tonometer
and
if
raised
checked
with
the
Gold-
mann
applanation
tonometer
with
the
slit
lamp.
Pupil
size
and
reactions
were
recorded,
as
was
the
general
health
of
the
exterior
of
the
eye.
After
dilatation
with
tropicamide
1%
the
optic
disc
vertical
cup
to
disc
ratio
BMJ
VOLUME
304
9
MAY
1992
Department
of
Preventive
Ophthalmology,
Institute
of
Ophthalmology,
London
EC1V
9EJ
R
P
L
Wormald,
lecturer
L
A
Wright,
research
assistant
P
Courtney,
lecturer
London
Ni
3NG
B
Beaumont,
general
practitioner
University
College
and
Middlesex
School
of
Medicine,
Whittington
Hospital,
London
N19
5NF
A
P
Haines,
professor
of
primary
health
care
Correspondence
to:
Mr
R
Wormald,
Academic
Unit
of
Ophthalmology,
Western
Ophthalmic
Hospital,
London
NWI
5YE.
BMJ
1992;304:1226-9
1226
was
recorded
and
the
macula
and
posterior
pole
examined
with
a
Volk
aspheric
90
dioptre
lens.
Lens
opacities
were
not
formally
graded
but
were
assessed
clinically
by
one
of
two
ophthalmologists
(RW
or
PC)
as
significant
or
otherwise.
When
indicated
the
retinal
periphery
was
examined
with
a
binocular
indirect
ophthalmoscope
and
28
dioptre
aspheric
lens.
Patients
found
to
have
ophthalmological
conditions
requiring
further
assessment
or
treatment,
or
both,
were
referred
to
a
local
eye
unit
(Moorfields
Eye
Hospital
or
the
Whittington
Hospital).
The
95%
confidence
intervals
around
estimates
of
prevalence
were
calculated
by
the
exact
method
for
small
proportions
and
the
normal
approximation
to
the
binomial
distribution
for
larger
ones.7
Prevalence
was
calculated
on
the
basis
of
subjects
examined,
although
some
data
were
available
on
the
prevalence of
visual
loss
among
non-respondents
from
the
practice
notes.
Results
There
was
no
significant
difference
in
the
propor-
tions
of
men
and
women
among
attenders
and
non-
attenders.
A
total
of
106
attenders
(51%)
were
aged
75
and
over.
PREVALENCE
OF
BLINDNESS
AND
VISUAL
IMPAIRMENT
Prevalence
estimates
are
summarised
in
the
table
with
95%
confidence
intervals
stratified
by
age
and
sex.
There
were
two
blind
participants
(World
Health
Organisation
criteria),
one
of
whom
was
registered.
Both
were
over
80.
Another
(also
over
80)
was
registered
blind
but
had
vision
better
than
3/60
in
one
eye.
Eight
subjects
fulfilled
the
American
criteria
of
legal
blind-
ness.
Low
vision
by
WHO
criteria
is
visual
acuity
less
than
6/18
in
either
eye:
16
subjects
had
low
vision
(7
7%;
95%
confidence
interval
4
5%
to
12-2%).
Visual
impairment
by
American
standards
is
best
acuity
of
less
than
6/12
(20/40)
and
better
than
6/60,
and
there
were
22
such
subjects
(10-6%;
95%
confidence
interval
6-9%
to
14-8%).
Eight
of
the
16
subjects
with
low
vision
(less
than
6/18)
were
known
by
their
doctor
to
have
an
eye
complaint.
Of
the
17
subjects
who
were
examined
in
their
homes,
seven
(41%;
95%
confidence
interval
18-4%
to
67-1%)
had
low
vision
(by
WHO
criteria).
Multiple
disabilities
were
common
in
this
group
and
in
no
case
was
visual
impairment
the
sole
cause
of
being
house-
bound.
Fifty
six
subjects
(27%;
95%
confidence
interval
21
-0%
to
33-
1%)
would
probably
have
benefited
from
a
new
refraction.
Among
these,
two
(11%)
cases
of
low
vision
(WHO
criteria)
were
aucounted
for
by
uncor-
rected
refractive
error,
and
21
(49%)
cases
of
visual
impairment
(American
criteria)
were
similarly
caused.
NON-RESPONDENTS
Eighty
one
people
did
not
participate
in
the
study.
Eleven
refused
saying
they
had
no
eye
problems
or
had
recently
seen
an
optician.
The
practice
notes
of
the
remaining
70
non-attenders
were
examined
for
any
comment
on
vision
or
attendance
to
an
eye
specialist.
It
was
apparent
from
the
notes
that
23
were
too
ill
to
participate
in
the
study.
Fifteen
non-respondents
were
already
under
the
care
of
an
eye
specialist
(19%),
one
person
was
registered
blind,
another
was
registrable,
and
one
was
registered
partially
sighted.
Four
were
visually
impaired
by
WHO
criteria
(prevalence
4-9%;
95%
confidence
interval
1-4%
to
12-2%).
There
were
eight
subjects
with
low
vision
known
to
the
practice
among
the
respondents
(3
9%;
1-7%
to
7-5%).
These
proportions
did
not
differ
significantly.
There
were
four
known
patients
with
glaucoma
receiving
treatment
who
did
not
attend
(prevalence
4
9%;
95%
confidence
interval
1-4%
to
12
2%).
The
remaining
32
had
no
comment
on
eye
health
or
significant
medical
problem
recorded
in
the
practice
notes.
PATHOLOGICAL
CAUSES
OF
BLINDNESS,
LOW
VISION,
AND
VISUAL
IMPAIRMENT
Blindness
in
one
woman
was
due
to
aging
maculo-
pathy
with
bilateral
disciform
degeneration.
The
second
had
thrombotic
glaucoma
in
one
eye
and
glaucoma,
pseudophakia,
and
disciform
degeneration
in
the
other.
Four
of
the
six
further
cases
found
with
vision
less
than
6/36
were
due
to
cataract.
The
fifth
was
due
to
ischaemic
optic
neuropathy
and
the
sixth
to
optic
atrophy.
In
cases
where
more
than
one
disorder
was
present,
such
as
cataract
or
mild
aging
changes
at
the
macula
(for
example,
drusen
or
minor
changes
in
the
retinal
pigment
epithelium),
a
clinical
decision
was
made
by
RPLW
in
order
to
determine
the
main
cause
of
blindness
or
visual
impairment.
Cataract
accounted
for
the
majority
of
cases
of
low
vision
and
visual
impairment
(see
figure).
The
preval-
ence
of
cataract
in
the
total
sample
tested,
as
defined
by
lens
opacity
reducing
the
vision
to
less
than
6/18,
was
5-8%
(95%
confidence
interval
3
0%
to
9
9%).
In
the
65-74
age
group
the
prevalence
of
cataract
was
1%
(0
03%
to
5
8%),
increasing
to
10-4%
in
the
group
aged
75
and
over
(4-6%
to
16-2%).
There
were
12
aphakic
participants
(5-8%;
95%
confidence
interval
3
0%
to
9
9%),
10
with
lens
implants
and
two
without.
Aging
maculopathy
caused
reduced
vision
(less
than
6/18)
in
the
better
eye
in
three
people
over
75,
giving
a
prevalence
of
2-8%
(95%
confidence
interval
0-5%
to
8
0%)
for
that
age
group
and
an
overall
prevalence
of
1-4%
(0-3%
to
4-2%)
(figure).
The
prevalence
of
Prevalence
of
blindness
and
visual
impairment
by
different
definitions
and
age
group
Age
65-74
years
Age
¢75
years
Total
95%
95%
95%
Prevalence
Confidence
Prevalence
Confidence
Prevalence
Confidence
Definition
No
(%)
interval
No
(%)
interval
No
(%)
interval
Males
and
females:
Blind
<3/60
0/101
00
2/106
1-9
0
2
to
6-6
2/207
1-0
0-15
to
3-4
Blind
<6/36
1/101
0.99
0
03
to
5-4
7/106
6-6
2-7
to
13-1
8/207
3-9
1-7
to
7
5
4/80*
5-0
1-4to
12-3
Low
vision
<6/18
1/101
0.99
0-03
to
5
4
15/106
14-2
7
5
to
20-8
16/207
7-7
4
5
to
12-2
Visual
impairment
<6/12
to
>6/60
1/101
0.99
0
03
to
5
4
21/106
19-8
12-2
to
27-4
22/207
10-6
6-4
to
14-8
Males:
Blind
<3/60
0/38
0
0
0/28
0
0
0/66
0 0
Blind
<6/36
0/38
0
0
1/28
3.57
0
09
to
18-4
1/66
1-5
4-3
to
8-2
1/24*
4-17
0-1
to21-1
Low
vision
<6/18
0/38
0
0
3/28
10-7
2-3
to
28-2
3/66
4-5
0-95
to
12
7
Visualimpairment<6/12to>6/60
0/38
0
0
6/28
21-4
8-3to40-9
6/66
9-1
3-4to
18-7
Females:
Blind
<3/60
0/63 0
0
2/78
2-6
0
3
to8-9
2/141
1-4
0-2
to5
0
Blind
<6/36
1/63
1
59
0
04
to
8
5
6/78
7-7
2-9
to
16
0
7/141
5-0
2-0
to
10-0
3/56*
5
4
1-1
to
14-9
Low
vision
<6/18
1/63
1-59
0
04
to
8
5
12/78
15-4
8-2
to
25
3
13/141
9-2
5
0
to
15-3
Visualimpairment<6/12to>6/60
1/63
1-59
0-04to8
5
15/78
19-2
11-2
to29-7
16/141
11-3
6-1
to
16-6
*Age
75-84
years.
BMJ
VOLUME
304
9
MAY
1992
1227
Visual
impairment
K
-
(A
miT
rican
criteria)
&~~~~~
~~~atro
':',;
$b
,
.,~
,G
i->
'}
j
,11/
Causes
of
low
vision
and
visual
impairment
clinically
obvious
predisposing
changes
at
the
macula
associated
with
visual
loss
in
at
least
one
eye
was
higher
(25%).
Three
new
cases
of
glaucoma
were
confirmed
in
the
sample
as
well
as
six
known
cases
in
people
who
were
already
receiving
treatment,
giving
an
overall
preval-
ence
of
4-4%
(95%
confidence
interval
2-0%
to
8
0%).
All
but
two
of
these
patients
were
over
75
years
of
age.
Diabetic
retinopathy
was
found
in
only
two
subjects
(1%;
0-1%
to
3
4%),
both
under
75.
There
were
14
known
diabetics
in
the
sample.
Many
other
minor
abnormalities
of
the
eyes
and
adnexa
were
found.
Only
68
participants
(33%)
had
an
entirely
normal
examina-
tion
result.
Thirty
three
people
were
referred
to
the
hospital
eye
service
(16%),
and
15
of
these
had
previously
been
under
the
care
of
an
eye
specialist,
but
only
three
had
active
hospital
numbers.
Although
only
three
of
the
13
referrals
for
suspected
glaucoma
could
be
confirmed
as
cases
(two
normal
tension
cases,
one
chronic
angle
closure
glaucoma),
all
but
one
were
followed
up
for
further
evaluation.
Other
referrals
were
for
cataract
(12),
lid
or
lacrimal
conditions
(four),
diabetic
retino-
pathy
(one,
newly
diagnosed),
branch
vein
occlusion
(one),
and
blind
or
partial
sight
registration
(two).
Discussion
The
response
rate
of
72%
in
this
survey
compared
favourably
with
the
Framingham
eye
study5
(67%
response)
but
was
less
than
in
some
other
studies.4810
When
the
Framingham
data
were
compared
with
ours
for
the
prevalence
of
any
lens
opacity
reducing
vision
to
less
than
6/12
(not
including
aphakia)
the
prevalence
in
London
was
significantly
higher
in
subjects
aged
75
and
over
(age
65-74:
Framingham
0
7%
(95%
con-
fidence
interval
0'3%
to
1-5%),
London
2%
(0-24%
to
7
0%);
age
-75:
Framingham
3-8%
(2-1%
to
6
2%),
London
16-4%
(9
0%
to
23
0%);
overall:
Framingham
1
7%
(1-0%
to
2
6%),
London
9-2%
(5
6%
to
14-0%)).
The
prevalence
of
aphakia
was
very
similar
in
the
two
studies
(4-2%
and
5-8%
with
overlapping
con-
fidence
intervals).
Three
per
cent
of
the
sample
in
a
Finnish
study
were
aphakic,9
but
the
author
com-
mented
that
in
53%
of
the
sample
with
vision
reduced
to
6/18
or
less
the
reduction
was
attributable
to
cataract.
This
compares
with
63%
in
our
study
when
using
the
same
criteria
(figure).
Though
the
overall
prevalences
of
cataract
were
similar
in
these
study
populations,
the
prevalence
of
visually
disabling
cataract
was
higher
in
our
sample,
especially
for
people
aged
75
and
over.
There
is
evidence
that
socioeconomic
status
is
a
determinant
of
cataract
risk,"
professional
classes
being
protected.
This
might
to
some
extent
explain
the
difference
between
our
sample
and
that
in
Framing-
ham.
Visual
impairment
was
common
(7/17
cases;
41%)
among
the
housebound.
Comparisons
of
prevalence
for
glaucoma
and
aging
maculopathy
were
more
difficult
because
of
differences
in
definition.
However,
there
seemed
to
be
no
statistic-
ally
significant
differences
for
glaucoma
between
this
study,
the
Framingham
study,
and
the
Finnish
study.9
In
the
Finnish
study
10%
of
visual
loss
to
a
level
of
6/18
or
less
was
due
to
aging
maculopathy
alone
whereas
we
attributed
20%
of
the
visual
loss
by
the
same
criteria.
However,
numbers
were
small
and
the
Finnish
study
did
not
attribute
principal
cause.
In
many
cases
more
than
one
cause
for
visual
loss
was
given.
Only
1%
of
visual
loss
was
attributed
to
aging
maculopathy
in
the
Framingham
study.
A
notable
difference
between
this
study
and
the
Finnish
one
was
the
number
of
participants
previously
or
currently
being
seen
by
an
eye
specialist.
Only
25%
of
the
participants
in
Turku,
Finland,
had
never
seen
an
ophthalmologist
compared
with
78%
in
this
study.
Also,
14
(11%)
of
those
who
had
never
seen
an
ophthalmologist
required
referral
but
only
one
of
these
referrals
was
for
cataract.
A
much
higher
proportion
of
participants
in
this
study
were
referred
for
cataract.
Access
to
an
ophthalmologist
is
clearly
easier
in
Finland
and
the
difference
may
also
be
due
to
greater
ophthalmological
involvement
in
routine
eye
testing
in
Finland.
The
number
of
ophthalmologists
per
head
of
popu-
lation
is
much
lower
in
the
United
Kingdom
than
the
average
for
European
Community
countries
(1/100
000
compared
with
1/20000)
(College
of
Ophthalmolo-
gists,
personal
communication).
However,
in
North
East
Thames
region
the
ratio
is
higher
than
the
national
average.
Recommendations
have
been
made
by
several
workers
over
the
past
few
years
for
screening
the
elderly
for
visual
disability.'2-'4
This
study
was
started
before
the
implementation
of
the
revised
general
practitioners'
contract,
which
requires
general
practi-
tioners
to
offer
yearly
health
screening
(including
assessment
of
vision
and
hearing)
to
everyone
aged
75
and
over
on
their
practice
lists.'3
An
assessment
has
been
recommended
based
on
the
ability
to
read
newspapers
and
recognise
faces
across
a
room.
13
Though
this
may
be
adequate
for
near
vision,
it
may
miss
people
who
have
remediable
loss
of
distance
acuity.
It
is
possible
to
recognise
familiar
people
with
remarkably
few
visual
clues.
The
Sonksen-Silver
3
m
visual
acuity
chart
is
easy
to
use
and
well
suited
to
use
in
primary
care.
A
3
m
length
of
cord
can
be
used
to
ensure
that
the
distance
is
correct.
Evaluation
of
the
red
reflex
of
the
eye
is
also
an
easy
test
and
quickly
performed.
It
will
identify
opacities
of
the
ocular
media,
which
most
frequently
will
be
due
to
cataract.
The
relatively
high
prevalence
of
refractive
error
suggests
the
need
for
assessment
by
an
optometrist.
However,
charges
for
sight
tests,
which
came
into
effect
before
the
study,
are
probably
a
disincentive
for
the
elderly,
especially
those
who
receive
only
a
state
pension,
to
make
use
of
optometrist
services.
This
study
suggests
that
there
is
a
considerable
amount
of
undetected
ocular
disease
and
potentially
remediable
disability
in
the
community.
Only
half
the
visually
disabled
subjects
were
known
to
their
doctors
to
have
visual
problems
before
the
survey.
We
thank
Dr
S
Hunt,
Dr
D
Davidson,
and
Dr
M
Cripwell
for
allowing
us
to
study
their
patients;
Drs
Daniel
Pauleikhoff
and
John
Wu,
who
helped
in
the
clinics;
Miss
Clare
Davey,
consultant
ophthalmologist,
for
reviewing
patients
referred
to
Whittington
Hospital
and
Professors
Bird
and
Jay,
whose
clinics
dealt
with
referrals
to
Moorfields
Eye
Hospital;
and
the
orthoptists
and
optometrists
who
participated
in
the
study.
We
also
acknowledge
the
assistance
given
by
Dr
Angela
12BMJ
VOLUME
304
9
MAY
1992
Lo*
vision
(WI4Ocrer);
1228
Reidy,
Norman
Dudley,
and
the
staff
of
the
Drovers
Day
Centre.
We
particularly
thank
Mr
J
Atwill,
house
governor
of
Moorfields
Eye
Hospital,
for
his
support
from
the
outset
of
the
study.
I
Brennan
ME,
Knox
EG.
An
investigation
into
the
purposes,
accuracy,
and
effective
uses
of
the blind
register
in
England.
Br
J
Prev
Soc
Med
1973;27:
154-9.
2
Cullinan
TR.
The
epidemiology
of
vtsual
disability:
studies
of
visually
disabled
people
in
the
community.
Canterbury:
Health
Services
Research
Unit,
1977.
(Report
28.)
3
Office
of
Population
Censuses
and
Surveys.
Surveys
of
disability
in
Great
Britain.
Report
I.
The
prn
valence
of
disability
among
adults.
London:
HMSO,
1988.
4
Gibson
JM,
Rosenthal
AR,
Lavery
J.
A
study
of
the
prevalence
of
eye
disease
in
the
elderly
in
an
English
community.
Trans
Ophthalmol
Soc
UK
1985;104:
196-203.
5
Framingham
eye
study
monograph.
Surv
Ophthalmol
l980;24(suppl):335-610.
6
Pizzarello
LD.
The
dimensions
of
the
problem
of
eye
disease
among
the
elderly.
Ophthalmology
1987;94:1
191-5.
7
Gardner
SB,
Winter
PD,
Gardner
MJ.
Confidence
interval
analvsis.
London:
British
Medical
Journal,
1989.
8
Martinez
GS,
Campbell
AJ.
Reinken
J,
Allan
BC.
Prevalence
of
ocular
disease
in
a
population
study
of
subjects
65
years
old
and
older.
Am
7
Ophthalmol
1982;94:
181
-9.
9
Hakkinen
L.
Vision
in
the
elderly
and
its
use
in
the
social
environment.
Scand
J
Soc
Med
1984;suppl
35.
10
Tielsch
JM,
Sommer
A,
Witt
K,
Katz
J,
Royall
RM.
Blindness
and
visual
impairment
in
an
American
urban
population:
the
Baltimore
eye
survey.
Arch
Ophthalmol
1990;108:286-90.
11
Leske
MC,
Chylack
LT,
Wu
S-Y.
The
lens
opacity
case-control
study.
Risk
factors
for
cataract.
Arch
Ophthalmol
1991;109:244-5
1.
12
Hitchings
RA.
Visual
disabilitv
and
the
elderly.
BMJ7
1989;298:1126-7.
13
Freer
CB.
Screening
the
elderly.
BMJ
1990;300:1447-8.
14
Sanderson
D.
Ocular
screening
for
the
elderly.
The
Canadian
Nurse
1986
Feb:
19-20.
(Accepted
18
Februarv
1992)
Effect
of
unexpected
demolition
announcement
on
health
of
residents
D
S
Halpern,
J
Reid
St
John's
College,
Cambridge
CB2
ITP
D
S
Halpern,
benefactors'
research
student
Halton
Health
Authority,
Runcorn
WA7
4TH
J
Reid,
consultant
in
health
medicine
Correspondence
to:
Mr
D
S
Halpern.
BMJ
1992;304:1229-30
Abstract
Objective-To
examine
the
impact
of
an
un-
expected
announcement
of
the
demolition
of
a
housing
estate
on
the
health
of
the
area's
residents.
Design-Study
of
general
practitioner
consulta-
tion
rates
of
the
estate's
residents
after
the
announce-
ment
compared
with
those
of
other
areas
and
with
those
of
the
previous
year.
Setting-General
practices
in
Runcorn,
Cheshire.
Patients-
17
000
patients
on
lists
of
the
two
group
practices
serving
the
estate
and
surrounding
area.
Main
outcome
measure-Relative
weekly
consulta-
tion
rates
with
general
practitioners.
Results-The
mean
adjusted
odds
ratio
for
con-
sultation
was
1-12
(SD
0.12)
when
demolition
was
expected
and
0X877
(0.05)
when
it
was
not
(t=5-94,
p<O0OOl).
The
difference
remained
after
the
adjust-
ment
for
the
fall
in
the
estate's
population
was
removed
(t=3-7,
p<0-01).
Conclusion-Announcement
of
the
estate's
demolition
adversely
affected
residents'
health.
Introduction
On
21
February
1989
the
decision
to
demolish
the
Southgate
housing
estate,
Runcorn,
a
total
of
1300
dwellings,
was
announced.
The
surprise
decision
dismissed
alternative
plans
for
the
estate's
refurbish-
ment
that
had
been
prepared
by
a
local
housing
association
in
consultation
with
the
residents.
The
development
corporation
sent
letters
to
the
estate's
2000
residents
informing
them
of
the
decision.
During
the
next
eight
weeks
there
was
vigorous
resident
activity
and
protest,
culminating
in
an
apparently
successful
meeting
with
the
under
secretary
of
state
(on
14
April
1989).
However,
within
four
weeks
the
original
decision
to
demolish
was
reconfirmed.
Anecdotal
experience
of
distressed
residents
was
apparent
to
the
local
public
health
physician
(JR),
who
was
participating
in
medical
rehousing,
and
we
conducted
this
study
to
quantify
the
effect
on
residents'
health.
Subjects
and
methods
We
studied
the
records
of
general
practitioner
consultations
for
the
first
six
months
of
1989
and,
for
comparison,
1988,
to
determine
changes
in
consultation
patterns.
We
manually
checked
the old
appointment
sheets
of
the
two
practices
that
served
the
estate
and
the
immediately
surrounding
area.
Each
consultation
was
classified
according
to
the
patient's
home
address
(Southgate
or
non-Southgate)
to
give
weekly
numbers
of
consultations.
The
weekly
numbers
of
consultations
by
Southgate
residents
(n)
were
then
expressed
as
a
ratio
of
consultations
by
non-Southgate
residents
(N).
The
ratio
(n/N)
controlled
for
spurious
fluctuations
in
the
absolute
numbers
of
patients
seen
in
any
given
week.
For
example,
seasonal
variations
in
consulta-
tions
would
cause
fluctuations
in
the
absolute
number
of
Southgate
and
other
patients
seen,
but
leave
the
ratio
largely
unaffected.
Similarly,
if
one
or
two
of
the
practices'
pine
doctors
happened
to
be
away
this
might
reduce
the
absolute
numbers
of
patients
seen
but
not
the
ratio.
(No
general
practitioner
specialised
in
patients
from
the
Southgate
estate.)
Hence,
the
ratio
provided
an
unbiased
estimate
of
the
relative
rate
of
Southgate
residents'
consultations
with
general
practi-
tioners
over
the
period
studied-that
is,
it
provided
a
good
measure
of
estate
specific
variation.
The
ratio
would
not
account
for
some
estate
specific,
seasonal
variation.
For
example,
interactions
between
the
time
of
year
and
the
consultation
rate
might
occur
due
to
the
relative
affluence
of
residents
(poorer
residents
being
less
able
to
heat
their
homes
in
winter).
Similarly,
if
an
estate
had
an
unusually
large
(or
small)
number
of
children
or
elderly
people,
and
if
these
groups
were
differentially
affected
by
variations
in
rates
of
illness,
this
could
introduce
estate
specific
variation.
This
was
controlled
for
by
comparing
con-
sultations
not
only
with
those
of
other
estates
in
the
same
week
but
also
with
the
rate
in
the
same
week
in
the
previous
year.
This
gave
the
formula
(n,/N1)/
(n2/N2),
the
odds
ratio
of
Southgate
consultations
to
non-Southgate
consultations,
1989
to
1988.
Finally,
we
corrected
for
changes
in
the
size
of
the
baseline
population.
The
number
of
occupied
units
on
the
estate
in
any
given
week
(U1)
was
derived
from
records
kept
by
the
officers
of
the
housing
association.
(The
housing
association
was
responsible
for
the
day
to
day
management
of
the
estate.)
This
allowed
the
calculation
of
an
adjustment
factor
(Ubase/UI),
where
Ubase
was
the
baseline
number
of
units
occupied
during
the past
year.
Delays
in
patients
registering
changes
in
address
were
accounted
for
by
offsetting
the
value
of
U,
by
three
weeks.
This
correction
factor
was
calculated
on
the
basis
of
information
from
the
practice
adminis-
trators.
This
led
to
a
final
formula
((n,/N,)/(n2/N2))
x
(Ubase,/U
1)
.
We
calculated
the
ratios
for
each
week
of
the
BMJ
VOLUME
304
9
MAY
1992
1229
